For the majority of non-vocal students with cognitive impairments, the most appropriate communication system involves graphics (simple symbols or picture). Unfortunately, other than multi-key or multi-symbol coding which are difficult to learn and recall, there are no rate enhancement techniques developed for these systems. The goal of this project is to develop and evaluate graphic based rate enhancement techniques that incorporate artificial intelligences and hypermedia technologies. The two technologies that will be developed and evaluated are graphic symbol prediction and a module to analyze the use of the communication system which can make recommendations to the therapist on how to recognize the language items to facilitate communication. A panel of experts will be identified to assist in developing the rules and protocols for rate enhancement. These features will then be incorporated into Communication Pages, which incorporates digital speech and dynamic visual displays in an easy-to-use hypermedia format. The graphic enhancement features will then be evaluated using a modified case study and multiple baseline approach involving four students. In addition, the features will be evaluated by therapists, parents, and significant others, and by experts in the field. Communication Pages will be marketed through existing commercial software distributors.